Skip navigation

Monthly Archives: March 2010

While on the subject of the apparently impossible, here’s another teasing puzzle that woodworkers can make to annoy their friends. It consists of 3 pieces: a cylinder, a symmetrical double cone and an inclined plane.

Surprisingly, placed together like this, there is no movement in either cylinder or cone. Wouldn’t you expect them to roll down the inclined plane?

The cylinder does, of course, roll down the plane but to take the photograph, I’ve used a ruler as a chock to prevent this happening.

The cone, on the other hand, has an inexplicable tendency to roll up hill. Once again, I’ve used a ruler as a chock to prevent it doing so.

These still photographs don’t really convey the anti-gravity properties of the double cone. For a more convincing demonstration, have a look at this video on YouTube.

Or, in case the link doesn’t work, paste this url into your browser:


As you can see, I’m working on a cello at the moment. But I’ve written about cello making before and, rather than repeat myself, I thought that I’d show a few wooden constructions that have amused me recently.

First, an old favourite – but one that people who don’t know the secret find seriously puzzling – the captive screw. There a trick to its manufacture, of course, and, if you can’t work out how it’s done, this YouTube video explains.

Rather more sophisticated is this apparently impossible double dovetail. There’s no trick here and the joint comes apart with ease. It’s just that the geometry of the joinery isn’t what one assumes it to be at first sight. The joint is occasionally useful. Roy Underhill describes an application in his book, The Woodwright’s Guide (ISBN978-0-8078-5914-8), where it’s employed to join the front legs to the top of a work bench.

The triple dovetail below is an ingenious puzzle that I found in Edwin Wyatt’s book, Wonders in Wood (ISBN 0-941936-40-6). Wyatt says that it was invented by someone called A B Cutler and published in a magazine, Industrial Arts and Vocational Education in 1930. It has no practical application as far as I know. The geometry of the joint is a variation on that of the double dovetail but it’s rather harder to make. Again, despite all appearances, the two pieces come apart easily – as you can see if you compare the positions of the ebony dots in the two photographs.

In a future post, I’ll show some photographs of the joints pulled apart.

%d bloggers like this: